@sujingshen: Spotify's best developer hasn't written a line of code since last December. Not laid off—just orchestrating Agents. Inside OpenAI it's even crazier: a 3-person team produced 1 million lines of code in 5 months, with zero manual writing. 99% of engineers' AI output is generated by Codex. …
Summary
The article points out that top developers at Spotify and OpenAI have stopped writing code manually, instead shifting to orchestrating AI Agents. Engineers are transitioning from coders to directors, focusing on constraint design, spec writing, harness construction, and output review.
View Cached Full Text
Cached at: 07/09/26, 09:40 AM
Spotify’s best developers haven’t written a single line of code since last December.
Not because they were laid off — they’ve been orchestrating agents.
Inside OpenAI, it’s even more extreme: a 3-person team produced 1 million lines of code in 5 months — 0 lines hand-written. 99% of the engineers’ AI output is generated by Codex.
Engineers are shifting from “people who write code” to “people who orchestrate agents.”
From actors to directors.
You don’t compete with agents on typing speed. Your new job is to get 10 agents running in the right direction at the same time.
The core skill stack has changed:
- Constraint design
- Spec writing
- Harness construction
- Output review
None of these is “writing code.” But each requires a higher system-level perspective than writing code ever did.
Full analysis + 10 data sources
Similar Articles
@0xMovez: Spotify's Chief Architect just showed how they ship 4,5K deployments /day with Claude at Anthropic stage 27-minutes. fr…
Spotify's Chief Architect revealed that over 99% of engineers use AI coding tools, shipping 4.5K deployments per day with Claude, and adoption surged after Claude Opus 4.5.
@Smartpigai: Anyone using AI for coding should bookmark this 113K stars open-source project: The Agency: a ready-made AI expert team. It organizes 232 specialized agents across 16 fields including frontend, backend, DevOps, product, design, marketing, security, testing, and more.
The Agency is a GitHub repository with 232 specialized AI agent prompts and workflows for roles like frontend, backend, DevOps, and more, designed to help developers use AI coding tools more effectively.
@laobaishare: GitHub steps in directly — from now on, no AI will write code blindly anymore. --- The newly released Spec Kit has soared to 95K stars in just a few days. The core idea is simple: make AI clearly specify what to do before touching any code. No more throwing a vague prompt and praying the agent doesn't blow up your project.
GitHub has released Spec Kit, a tool that forces AI to generate structured specifications before writing code, including understanding requirements, asking for missing details, organizing the project, and more. It significantly reduces AI-generated error-prone code and is compatible with 25+ AI agents.
@FakeMaidenMaker: The scariest thing about using an AI agent to write code is losing control: the agent runs wild, quality is inconsistent, you don’t know what stage it’s in, and it messes things up halfway through. AWS just open-sourced a set of development lifecycle workflow rules specifically designed for AI coding agents — AI-DLC — that make the agent…
AWS has open-sourced AI-DLC (AI-Driven Development Life Cycle), a set of development lifecycle workflow rules designed for AI coding agents to help developers control agent behavior and ensure quality. It supports multiple platforms including Claude Code, Cursor, and GitHub Copilot.
@billtheinvestor: 95K: GitHub just pushed the development boundary of AI Agents one step forward. This newly open-sourced system forces AI to complete a full specification (Specs) before writing code. Raking in 95K Stars within days, the most direct consequence is that AI is shifting from 'blind code writing' to 'thinking before executing.'
An open-source system that forces AI to complete a full specification before writing code appeared on GitHub, garnering 95K Stars within days, pushing AI from blind code writing to thinking before executing.